Abstract Brain-derived neurotrophic factor (BDNF) is a nervous system growth factor that enhances synaptic plasticity and regulates neuronal function. BDNF gene therapy for Alzheimer?s disease (AD) is a promising alternative to amyloid- and tau-targeted therapies: BDNF reduces neuronal degeneration and stimulates neuronal activity in rodent and non-human primate models of AD. Direct injection of an Adeno- Associated Virus (AAV) vector into entorhinal cortex mediates safe and long-lasting BDNF expression, and will soon begin human clinical trials. Although promising, intraparenchymal AAV-BDNF injection is invasive and treats only a small percentage of the cerebral cortex. Intrathecal administration of AAV9-BDNF to the cerebrospinal fluid could solve these problems by broadly treating the entire cortex from a single minimally invasive infusion. We recently reported that two hours of Trendelenburg positioning, in which the body lies supine on a reclining table with the head 30 below the feet, dramatically increases the strength and consistency of gene transfer to cerebral cortex after intrathecal AAV9 infusion in rats. More than 95% of transduced cells in cortex are neurons, and gene expression in off-target brain regions and spinal cord is minimal. This novel delivery method has strong potential for clinical treatment of AD. We propose systematic preclinical testing of intrathecal AAV9-BDNF gene therapy for AD. Aim 1 will test therapeutic efficacy by directly comparing intrathecal and intraparenchymal AAV9-BDNF infusion in a transgenic mouse model of AD and analyzing behavioral and anatomical outcomes. Aim 2 will test the safety of intrathecal AAV9-BDNF infusion at escalating doses and over prolonged treatment periods in the non-human primate. Aim 3 will enhance the specificity of intrathecal AAV9-BDNF therapy by testing cell-specific promoters to reduce or eliminate off- target gene expression. These studies aim to simplify delivery, enhance efficacy, and increase clinical feasibility of BDNF gene therapy for AD, and will support both upcoming clinical trials and preclinical development of new gene therapies for AD.